Catphan 604 Slice Thickness

[Jonny]

Hi,

I was wondering if someone could help me. I have two identical scans with a head protocol except that one was scanned with a 2mm slice thickness while the other was scanned with a 2.5mm slice thickness, these can be found here:  https://drive.google.com/drive/folders/1xN6yaYCoSkw6T1kagAXbl8vpDragRUQa?usp=sharing

Pylinac calculates the slice thickness of the 2mm scan as 1.778mm which is fine, however, it calculates the slice thickness of the 2.5mm scan as 1.578mm, smaller than the 2mm scan.

Looking at the images on ImageJ the wire is visibly wider on the 2.5mm scan and after a quick manual look at the FWHM it appears larger.

I was hoping someone might have some insight on this ?

Thank you very much for any replies.

John.

[James Kerns]

Hi John, 

   It looks like pylinac is finding the center of the 604 slightly off-center in the S-I direction. When running your images (in v3.0 FWIW) the analyzed image shows the ramp almost at the end. Note how far from centered it is. I'm wondering if there weren't some tweaks made to the 604 in production that made the HU cylinders longer than the ramp section. Anyway, the issue was that the FWHM wasn't being found correctly because it was at the edge (see image). I tweaked the offset of the 404 module and got an outcome of 2.19mm. You can see how to do that here, but I also made a gist to do exactly what I described: offset 404 (github.com)